the four ClosestPointOnSdf nodes seem to render unexpected results, away from the sdf
attached help file renders
03 40 29 17

is this node supposed to get the closest point on the surface to an arbitrary point? in that case you can calculate that yourself very easily. You already know the distance to that closest point because you’re working with a distance field, so you need to calculate only the derivative of the field (so you get the direction to the closest point from the arbitrary point), normalize and multiply that direction with the distance you already get from the DF and add that to your arbitrary point. boom you have the closest point to some error threshold depending how precise the direction calculation was.

It looks like that direction calculation has a flipped sign in that node or in your case for some reason.

Yes that’s what the node does (and how it works inside too).

Working fine here:

@ggml pls check your packs are up to date, pretty sure that will fix it, and if you still have the problem file an issue on the FieldTrip github repo (it’s super easy, no code or git stuff required).

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.